Partiality vector refinement systems and methods through sample probing

• US 10,430,817 B2
• Filed: 04/14/2017
• Issued: 10/01/2019
• Est. Priority Date: 04/15/2016
• Status: Active Grant

• Current Assignee: Walmart Apollo, LLC (WalMart Inc.)
• Original Assignee: Walmart Apollo, LLC (WalMart Inc.)
• Inventors: Wilkinson, Bruce W., Mattingly, Todd D.
• Primary Examiner(s): Mansfield, Thomas L
• Application Number:
  US15/487,882
• Publication Number:
  US 20170300944A1
• Time in Patent Office: 900 Days
• Field of Search: None
• CPC Class Codes: G06F 16/288 Entity relationship models
• CPC Class Codes: G06Q 30/0203 Market surveys; Market polls

Abstract

Some embodiments provide customer partiality vectorization refinement systems, comprising: a customer database storing a set of multiple customer partiality vectors for each of multiple different customers; a product database storing a set of multiple product partiality vectors for each of multiple different products; and a vectorized refinement control circuit configured to: identify, for a first customer, a multi-dimensional partiality vector target area defined by a limited range of partiality magnitudes and limited range of representative partiality directions; select a first product having at least a product partiality vector that is within a threshold alignment with the partiality vector target area, and cause the first product to be presented to the first customer; receive feedback associated with the first customer and corresponding to the first product; and adjust the partiality vector target area based on the feedback.

Claims

1. A retail shopping customer partiality vectorization refinement system, comprising:

a memory having stored therein;

a customer database storing at least a different set of multiple customer partiality vectors for each of multiple different customers wherein each of the customer partiality vectors has at least one of a magnitude and an angle that corresponds to a magnitude for the person associated with that partiality; and

a product database storing at least a different set of multiple product vectorized characterizations for each of multiple different products, wherein each of the vectorized characterizations indicates a measure regarding an extent to which a corresponding one of the products accords with a corresponding one of the multiple customer partiality vectors; and

a vectorized refinement control circuit operably coupled with the customer database and the product database, wherein the refinement control circuit;

identifies, for a first customer of the multiple customers, a multi-dimensional partiality vector target area defined within a multi-dimensional representative volume defined by a limited range of partiality magnitudes and limited range of representative partiality directions for a first customer partiality vector, wherein the partiality vector target area represents a multi-dimensional representation of an area in which an unknown actual magnitude and direction for the first customer partiality vector are predicted to lie;

selects a first product from the multiple different products having at least a first product vectorized characterization that is within a threshold alignment with the partiality vector target area, and causes the first product to be presented to the first customer;

receives, following the first product being presented to the first customer, feedback associated with the first customer and corresponding to the first product; and

adjusts the partiality vector target area based on the feedback;

wherein the customer database comprises a distributed database maintained across at least multiple different customer computing devices, and the refinement control circuit receives processing of at least the distributed database from a plurality of customer computing device control circuits located in the multiple different customer computing devices that are geographically distributed over a geographic area.

2. The system of claim 1, wherein the refinement control circuit further confirms, based on the feedback comprising at least one subsequent purchase of the first product by the first customer, a level of accuracy of the partiality vector target area;

and

reduces, in adjusting the partiality vector target area, the partiality vector target area.

3. The system of claim 2, wherein the refinement control circuit, in causing the first product to be presented to the first customer, causes a sample of the first product to be provided to the first customer.

4. The system of claim 2, wherein the refinement control circuit, in adjusting the partiality vector target area, reduces the range of partiality magnitudes defining the partiality vector target area.

5. The system of claim 4, wherein the refinement control circuit, in adjusting the partiality vector target area, adjusts the range of representative partiality directions defining the partiality vector target area.

6. The system of claim 1, wherein the refinement control circuit, in selecting the first product, tests an accuracy of the partiality vector target area by selecting the first product having the first product vectorized characterization that has the alignment that is a threshold vector outside of the partiality vector target area.

7. The system of claim 1, wherein the refinement control circuit, in selecting the first product, tests an accuracy of the partiality vector target area by selecting the first product having the first product vectorized characterization and costs the first customer a premium.

8. The system of claim 7, wherein the premium cost comprises a magnitude of a second product vectorized characterization of the first product that is a premium threshold less than a magnitude of a second customer partiality vector of the first customers.

9. A method of refining retail shopping customer partiality vectors, comprising:

accessing a customer database storing at least a different set of multiple customer partiality vectors for each of multiple different customers wherein each of the customer partiality vectors has at least one of a magnitude and an angle that corresponds to a magnitude for the person associated with that partiality, and a product database storing at least a different set of multiple product vectorized characterizations for each of multiple different products, wherein each of the vectorized characterizations indicates a measure regarding an extent to which a corresponding one of the products accords with a corresponding one of the multiple customer partiality vectors;

identifying, for a first customer of the multiple customers, a multi-dimensional partiality vector target area defined within a multi-dimensional representative volume defined by a limited range of partiality magnitudes and limited range of representative partiality directions for a first customer partiality vector, wherein the partiality vector target area represents a multi-dimensional representation of an area in which an unknown actual magnitude and direction for the first customer partiality vector are predicted to lie;

selecting a first product from the multiple different products having at least a first product vectorized characterization that is within a threshold alignment with the partiality vector target area, and causing the first product to be presented to the first customer;

receiving, following the first product being presented to the first customer, feedback associated with the first customer and corresponding to the first product;

adjusting the partiality vector target area based on the feedback and

implementing at least a portion of the customer database as a distributed customer database maintained across at least multiple different customer computing devices, and causing processing of at least the customer partiality vectors through a plurality of customer computing device control circuits located in the multiple different customer computing devices that are geographically distributed over a geographic area.

10. The method of claim 9, further comprising:

confirming, based on the feedback comprising at least one subsequent purchase of the first product by the first customer, a level of accuracy of the partiality vector target area; and

reducing, in adjusting the partiality vector target area, the partiality vector target area.

11. The method of claim 10, wherein the causing the first product to be presented to the first customer comprise causing a sample of the first product to be provided to the first customer.

12. The method of claim 10, wherein the adjusting the partiality vector target area comprises reducing the range of partiality magnitudes of the first customer partiality vector.

13. The method of claim 12, wherein the adjusting the partiality vector target area comprises adjusting the range of representative partiality directions of the first customer partiality vector.

14. The method of claim 9, further comprising:

testing an accuracy of the partiality vector target area comprises implementing the selecting of the first product such that the first product has the first product vectorized characterization that has the alignment that is a threshold vector outside of the partiality vector target area.

15. The method of claim 9, further comprising:

testing an accuracy of the partiality vector target area comprises implementing the selecting of the first product such that the first product has the first product vectorized characterization and costs the first customer a premium.

16. The method of claim 15, wherein the selecting the first product comprises selecting the first product having the premium cost comprising a magnitude of a second product vectorized characterization of the first product being a premium threshold less than a magnitude of a second customer partiality vector of the first customer.